Neuritic plaques and neurofibrillary tangles are the hallmark pathologies of Alzheimer's disease (AD), but the presence individual tangles or plaques is not sufficient to predict degeneration; 45% of the elderly with plaques and tangles lack cognitive loss or dementia. We have recently identified a new type of molecular pathology in AD that derives from the aggregation of RNA binding proteins (RBPs), forming RNA-protein complexes, which are termed RNA granules. Microtubule associated protein tau (MAPT) binds to RBPs, co-localizes with RBPs in RNA granules, and RBPs increase MAPT misfolding /aggregation. Importantly, reducing the RBP TIA1 delays progression of tauopathy, despite increased MAPT aggregation. We hypothesize that variation in the composition of MAPT complexes (soluble or insoluble) and RNA granule complexes represent critical determinants of the molecular heterogeneity of AD and other tauopathies, and identify particular pathways that uniquely contribute to each type of disease. We will apply systems biology approaches that integrate information from proteomics and RNA metabolism to identify key proteins in each complex that are associated with neurodegeneration, and then test the roles of these proteins/genes experimentally. Throughout this proposal we will use unbiased studies (e.g., proteomic and RNAseq) combined with the systems biology algorithms to model context-dependent information flows to identify key molecular interactions and pathways regulating pathology, neurodegeneration and neuroprotection. Aim 1 will determine whether the RBP TIA1 directs the biochemical and functional properties of MAPT aggregation. We have discovered that reducing the RBP TIA1 delays disease progression in PS19 P301S MAPT mice despite producing more aggregation. We will elucidate the mechanisms by which TIA1 reduction produces neuroprotection using both in vitro molecular studies, and use unbiased ?omic? studies (mass spectrometry and RNAseq). We will apply the systems biology algorithms to quantify key gene-gene interactions and pathways, and identify those pathways that parallel the human condition. Aim 2 will determine how MAPT and RBP complexes vary with cognitive decline in humans. We will use mass spectrometry and RNAseq to determine how the composition of complexes of MAPT, TIA1 and other key RBPs varies among human cases exhibiting neuritic plaques and neurofibrillary tangles with or without cognitive decline. Aim 3 will determine whether RBPs direct the strain of MAPT and resulting pathologies that are propagated in vitro and in vivo. We will characterize propagation of tauopathy for MAPT aggregates from PS19, PS19xTIA1+/--mice, as well as human cases exhibiting MAPT pathology with and without cognitive decline. The resulting mice will be analyzed as described in Aim 1.